Asymptotic Theory and Statistical Inference for the Sample Problems with Heavy-Tailed Data using the Functional Empirical Process
Afrika Statistika · 01.01.2025 · 20 (1) · 4289-4318
DOI: 10.16929/as/2025.4289.347 ↗Kurz gefasst
Fachartikel in der Zeitschrift Afrika Statistika, online erschienen am 01.01.2025. Verfasst von Tchilabalo Abozou Kpanzou, Abdoulaye Camara und 1 weiteren Autor:innen.
Abstract (Originalsprache)
We propose the Trimmed Functional Empirical Process (TFEP) for statistical inference under heavy-tailed or skewed distributions, for which classical moments may be infinite. Unlike the classical Functional Empirical Process (FEP), which requires finite second moments and fails for distributions such as Pareto, Cauchy, or Student-t with low degrees of freedom, the TFEP stabilizes the empirical process by trimming extreme observations. We establish the weak convergence of the TFEP under mild conditions and derive asymptotic distributions for one-sample and two-sample problems. Monte Carlo simulations and an application to Senegalese income data demonstrate the robustness and effectiveness of the proposed approach.
Zitieren
APA
Camara, T. A. K. A., & Lo, C. S. D. M. D. D. G. S. (2025). Asymptotic Theory and Statistical Inference for the Sample Problems with Heavy-Tailed Data using the Functional Empirical Process. Afrika Statistika, 20(1), 4289-4318. https://doi.org/10.16929/as/2025.4289.347
Vancouver
Camara TAKA, Lo CSDMDDGS. Asymptotic Theory and Statistical Inference for the Sample Problems with Heavy-Tailed Data using the Functional Empirical Process. Afrika Statistika. 2025;20(1):4289-4318. doi:10.16929/as/2025.4289.347
BibTeX
@article{Camara2025,
author = {Camara, Tchilabalo Abozou Kpanzou, Abdoulaye and Lo, Camara, Saliou Diouf, Moumouni Diallo Diallo, Gane Samb},
title = {Asymptotic Theory and Statistical Inference for the Sample Problems with Heavy-Tailed Data using the Functional Empirical Process},
journal = {Afrika Statistika},
year = {2025},
volume = {20},
number = {1},
pages = {4289--4318},
doi = {10.16929/as/2025.4289.347}
}
EUVIMED